Electroencephalogram decision system, electroencephalogram decision method, program, and non-transitory storage medium

Abstract

An electroencephalogram decision system determines, when finding electroencephalogram information representing an electroencephalogram obtained by an electrode unit placed on a region of interest that forms part of a subject's head satisfying a predetermined condition, that a target electroencephalogram, which is an electroencephalogram with a characteristic variation, should have been produced. The predetermined condition is selected from of detection conditions including a first type of condition and a second type of condition. The first type of condition specifies that a decision should be made, based on a component falling within a single frequency band included in the electroencephalogram represented by the electroencephalogram information, whether or not the target electroencephalogram has been produced. The second type of condition specifies that a decision should be made, based on components falling within multiple different frequency bands included in the electroencephalogram represented by the electroencephalogram information, whether or not the target electroencephalogram has been produced.

The invention claimed is: 
     
       1. An electroencephalogram decision system comprising:
 an acquisition unit configured to acquire electroencephalogram information, the electroencephalogram information representing an electroencephalogram obtained by an electrode unit, the electrode unit configured to be placed on a region of interest that forms part of a subject's head; 
 an analysis unit configured to calculate each of powers of first and second frequency bands included in a frequency spectrum of the electroencephalogram represented by the electroencephalogram information acquired by the acquisition unit, the first and second frequency bands being different from each other; 
 a detection unit configured to determine, based on a decision condition, whether or not the electroencephalogram information represents a target electroencephalogram which is an electroencephalogram with a characteristic variation; 
 a processing unit; and 
 at least one of (a) an exercise assisting device configured to support the subject in his or her voluntary movement by applying at least one of a mechanical stimulus or an electrical stimulus to the subject in coordination with the determination by the detection unit or (b) a display unit configured to display an image for guiding his or her voluntary movement based on the determination by the detection unit, 
 the decision condition being selected from a plurality of conditions, 
 the plurality of conditions including a first type of condition and a second type of condition, 
 the first type of condition determines that the acquired electroencephalogram information equals the target electroencephalogram when a power of a single one of the first and second frequency bands transitions from a rest range to an exercise range 
 the second type of condition determines that the acquired electroencephalogram information equals the target electroencephalogram when each of the powers of the first and second frequency bands transition from the rest range to the exercise range, 
 the processing unit being configured to select the decision condition from the plurality of conditions based on a rate of change of a border line between the rest range and the exercise range on a graph where one axis represents a power of the first frequency band and another one-axis represents a power of the second frequency band, 
 wherein, where the detection unit determines that the electroencephalogram information represents the target electroencephalogram, the at least one of the exercise assisting device or the display unit respectively supports the subject in his or her voluntary movement by applying the at least one of the mechanical stimulus or the electrical stimulus to the subject or displaying the image for guiding his or her voluntary movement. 
 
     
     
       2. The electroencephalogram decision system of  claim 1 , wherein
 the first frequency band is an a-wave frequency band, and 
 the second frequency band is a B-wave frequency band. 
 
     
     
       3. The electroencephalogram decision system of  claim 1 , wherein
 the plurality of conditions further includes a third type of condition specifying that a decision be made, based on a power of one or more frequency bands selected according to a healthy person's electroencephalogram, whether or not the electroencephalogram information represents the target electroencephalogram. 
 
     
     
       4. The electroencephalogram decision system of  claim 1 , wherein
 the rest range is a range in which each of the powers of the first and second frequency bands of the acquired electroencephalogram are able to fall when the subject puts his or her body at rest. 
 
     
     
       5. The electroencephalogram decision system of  claim 1 , wherein
 the exercise range is a range in which each of the powers of the first and second frequency bands of the acquired electroencephalogram are able to fall when the subject plans to do voluntary movement. 
 
     
     
       6. The electroencephalogram decision system of  claim 1 , wherein the processing unit is configured to, when finding the rate of change of the border line is negative, select the second type of condition. 
     
     
       7. The electroencephalogram decision system of  claim 6 , wherein
 the detection unit is configured to determine, by seeing each of the powers of the first and second frequency bands make a transition from the rest range to the exercise range, that the electroencephalogram information represents the target electroencephalogram according to the second type of condition. 
 
     
     
       8. The electroencephalogram decision system of  claim 1 , wherein
 the plurality of conditions further includes a third type of condition specifying that a decision be made, based on a power of one or more frequency bands selected according to a healthy person's electroencephalogram, whether or not the electroencephalogram information represents the target electroencephalogram, and 
 the processing unit is configured to select the third type of condition when finding the rate of change of the border line is zero or positive. 
 
     
     
       9. The electroencephalogram decision system of  claim 1 , wherein the rate of change of the border line between the rest range and the exercise range on the graph where the one axis represents the power of the first frequency band and the another axis represents the power of the second frequency band corresponds to a slope of a decision boundary, calculated by linear discriminant analysis (LDA) method, between the rest range and the exercise range on the graph where the one axis represents the power of the first frequency band and the another one-axis represents the power of the second frequency band. 
     
     
       10. An electroencephalogram decision method comprising:
 a first step of acquiring electroencephalogram information, the electroencephalogram information representing an electroencephalogram obtained by an electrode unit, the electrode unit being placed on a region of interest that forms part of a subject's head; 
 an analysis step of calculating each of powers of first and second frequency bands included in a frequency spectrum of the electroencephalogram represented by the electroencephalogram information acquired by the first step, the first and second frequency bands being different from each other; 
 a second step of making a decision, based on a decision condition, whether or not the electroencephalogram information represents a target electroencephalogram which is an electroencephalogram with a characteristic variation; and 
 at least one of (a) an exercise assisting step of supporting the subject in his or her voluntary movement by applying at least one of a mechanical stimulus or an electrical stimulus to the subject in coordination with the determination by the second step or (b) a displaying step of displaying an image for guiding his or her voluntary movement based on the determination by the second step, 
 the decision condition being selected from a plurality of conditions, 
 the plurality of conditions including a first type of condition and a second type of condition, 
 the first type of condition determining that the acquired electroencephalogram information equals the target electroencephalogram when a power of a single one of the first and second frequency bands transitions from a rest range to an exercise range 
 the second type of condition determining that the acquired electroencephalogram information equals the target electroencephalogram when each of the powers of the first and second frequency bands transition from the rest range to the exercise range 
 the electroencephalogram decision method further comprising a processing step of selecting the decision condition from the plurality of conditions based on a rate of change of a border line between the rest range and the exercise range on a graph where one axis represents a power of the first frequency band and another axis represents a power of the second frequency band, 
 wherein, where the second step making a decision that the electroencephalogram information represents the target electroencephalogram, the at least one of the exercise assisting step or the displaying step respectively supports the subject in his or her voluntary movement by applying the at least one of the mechanical stimulus or the electrical stimulus to the subject or displaying the image for guiding his or her voluntary movement. 
 
     
     
       11. The electroencephalogram decision method of  claim 10 , wherein the rate of change of the border line between the rest range and the exercise range on the graph where the one axis represents the power of the first frequency band and the another axis represents the power of the second frequency band corresponds to a slope of a decision boundary, calculated by linear discriminant analysis (LDA) method, between the rest range and the exercise range on the graph where the one axis represents the power of the first frequency band and the another axis represents the power of the second frequency band. 
     
     
       12. A non-transitory storage medium readable for a computer system including a processor and storing a program designed to cause the processor to carry out the electroencephalogram decision method of  claim 10 .